The present teachings relate generally to the field of buck converters, and more particularly, to switching control thereof.
These teachings relate to switching power supplies (switching converters). These devices are used to efficiently transform voltage and currents at one level to voltage and currents at a different level. Switching converters are particularly important when either high power or battery operation require high efficiency. Switching converters are pervasive throughout many consumer products such as cell phones, PDAs, personal computers, etc. A key feature of the switching power supply is its small size and low cost, which is achieved thru efficient design.
One of the ways to improve the size and cost of switching converters is to optimize the size of the external passive components. This is achieved by optimizing power device topology.
With the advent of deep sub-micron CMOS, power supplies with very low voltage, high tolerance and high currents are required. As a result, passive filter components have to be scaled to a very low impedance, and in particular the output capacitor is selected to be of high quality and large value. This capacitor dominates the size and cost of the switching converters for sub-micron CMOS. In general, a smaller the capacitor means lower cost.
There is a need for power device topologies that allow the output capacitor to be reduced.
Buck DC-to-DC converters are frequently used to provide lower operating voltages as required in various devices that use integrated circuits. Typically a much larger supply voltage is used with a narrow duty cycle to charge an output capacitor through an inductor, with accumulated inductor current maintaining the output voltage when not connected to the higher voltage supply during the off portion of the duty cycle.
This arrangement of having a higher supply voltage for charging the inductor current and a much smaller output voltage for discharging the inductor current causes the step load recovery to be asymmetrical. When the load current steps from the small value to near its maximum value, the voltage across the inductor is the supply voltage minus the output voltage. In this case, the relatively large supply voltage allows the inductor to be charged quickly, and the corresponding output voltage droop is minimized.
When the load current steps from near its maximum value to near its minimum value, the voltage across the inductor is just the output voltage. This relatively small voltage discharges the inductor slowly. This has the unfortunate result of allowing a large output voltage overshoot. This is an inherent limitation of the fundamental buck converter structure.
One method to avoid this problem employs discrete MOSFETs to allow the bulk diode of the synchronous rectifier FET to turn on for a short time. This increases the discharge voltage by the turn-on voltage of the bulk diode. Unfortunately this method is not typically available in the integrated circuit form. In integrated circuit power devices, it is desirable not to turn on the bulk diodes for latch-up reasons. Latch-up is self destructive in most integrated circuits and efforts are made to avoid it. In addition, the improvement in the discharge rate is only increased by the addition of a diode forward drop.
For the above reasons, it would be beneficial to provide improved methods and devices for discharging built up inductor current in buck converters when the converter experience is a substantial drop in load current being drawn there from, without using a larger capacitor to store the extra charge.